Jakub Grela

The increasing complexity of residential energy systems, driven by rising electricity demand, renewable energy integration, and dynamic pricing mechanisms, has intensified the need for intelligent load scheduling within Home Energy Management Systems (HEMSs). This paper presents a comprehensive literature review of optimization algorithms applied to residential load scheduling, based on an analysis of 78 peer-reviewed studies published between 2020 and 2025. The analysis reveals a clear shift from conventional deterministic optimization toward adaptive and data-driven approaches capable of operating in uncertain and dynamic environments. Metaheuristic methods are widely used for solving complex scheduling problems, while Machine Learning and Deep Learning (DL) techniques primarily support forecasting tasks related to energy demand and renewable generation. Reinforcement Learning (RL) and Deep Reinforcement Learning (DRL) approaches enable autonomous real-time decision-making, although challenges related to scalability, computational cost, and practical deployment remain unresolved. The review identifies hybrid architectures that combine forecasting, optimization, and control mechanisms as the most promising direction for future HEMS development. Finally, the paper highlights key research gaps, including limited real-world validation, insufficient consideration of physical infrastructure constraints, and the need for scalable distributed control frameworks for future smart grids and energy communities.

The residential energy sector contributes substantially to global energy-related emissions. Effective energy management requires an understanding occupant behavior through activity detection and habit identification. Recent advances in artificial intelligence (AI) and machine learning (ML) enable the automatic detection of user activities and prediction of energy needs based on historical consumption data. Non-intrusive load monitoring (NILM) facilitates device-level disaggregation without additional sensors, supporting demand forecasting and behavior-aware control in Home Energy Management Systems (HEMSs). This review synthesizes various AI and ML approaches for detecting user activities and energy habits in HEMSs from 2020 to 2025. The analyses revealed that deep learning (DL) models, with their ability to capture complex temporal and nonlinear patterns in multisensor data, achieve superior accuracy in activity detection and load forecasting, with occupancy detection reaching 95–99% accuracy. Hybrid systems combining neural networks and optimization algorithms demonstrate enhanced robustness, but challenges remain in limited cross-building generalization, insufficient interpretability of deep models, and the absence of dataset standardized. Future work should prioritize lightweight, explainable edge-ready models, federated learning, and integration with digital twins and control systems. It should also extend energy optimization toward occupant wellbeing and grid flexibility, using standardized protocols and open datasets for ensuring trustworthy and sustainability.

Indoor air quality (IAQ) is a critical determinant of health, comfort, and productivity, and is strongly connected to building energy demand due to the role of ventilation and air treatment in HVAC systems. This review examines recent applications of Artificial Intelligence (AI) and Machine Learning (ML) for IAQ prediction across residential, educational, commercial, and public environments. Approaches are categorized by predicted parameters, forecasting horizons, facility types, and model architectures. Particular focus is given to pollutants such as CO₂, PM2.5, PM10, VOCs, and formaldehyde. Deep learning methods, especially the LSTM and GRU networks, achieve superior accuracy in short-term forecasting, while hybrid models integrating physical simulations or optimization algorithms enhance robustness and generalizability. Importantly, predictive IAQ frameworks are increasingly applied to support demand-controlled ventilation, adaptive HVAC strategies, and retrofit planning, contributing directly to reduced energy consumption and carbon emissions without compromising indoor environmental quality. Remaining challenges include data heterogeneity, sensor reliability, and limited interpretability of deep models. This review highlights the need for scalable, explainable, and energy-aware IAQ prediction systems that align health-oriented indoor management with energy efficiency and sustainability goals. Such approaches directly contribute to policy priorities, including the EU Green Deal and Fit for 55 package, advancing both occupant well-being and low-carbon smart building operation.

This paper explores the development and optimization of a hybrid renewable energy system (HRES) integrated with a hybrid battery energy storage system (HBESS) to achieve energy autonomy for an e-Houseboat. The e-Houseboat is a floating residential unit equipped with advanced sustainable technologies, including photovoltaic panels, wind turbines, and a hybrid battery storage system consisting of lithium iron phosphate (LFP) and lead-acid batteries. The primary goal of this study was to design an energy-autonomous e-Houseboat capable of sustaining energy demands for at least one month without external power sources, regardless of the season. This study included a comprehensive analysis of energy generation potential from renewable sources across different European locations, detailed simulations of the energy storage system, and the development of energy management function for a houseboat automation system. The results demonstrate the feasibility of achieving the desired energy autonomy by leveraging the synergistic benefits of multiple energy storage technologies and optimizing energy management strategies. The experiment demonstrated that the implemented solutions enabled the facility to achieve energy autonomy for a period of 7 months.

In the context of the increasing integration of renewable energy sources (RES) and smart devices in domestic applications, the implementation of Home Energy Management Systems (HEMS) is becoming a pivotal factor in optimizing energy usage and reducing costs. This review examines the role of reinforcement learning (RL) in the advancement of HEMS, presenting it as a powerful tool for the adaptive management of complex, real-time energy demands. This review is notable for its comprehensive examination of the applications of RL-based methods and tools in HEMS, which encompasses demand response, load scheduling, and renewable energy integration. Furthermore, the integration of RL within distributed automation and Internet of Things (IoT) frameworks is emphasized in the review as a means of facilitating autonomous, data-driven control. Despite the considerable potential of this approach, the authors identify a number of challenges that require further investigation, including the need for robust data security and scalable solutions. It is recommended that future research place greater emphasis on real applications and case studies, with the objective of bridging the gap between theoretical models and practical implementations. The objective is to achieve resilient and secure energy management in residential and prosumer buildings, particularly within local microgrids.

In the context of the increasing integration of renewable energy sources (RES) and smart devices in domestic applications, the implementation of Home Energy Management Systems (HEMS) is becoming a pivotal factor in optimizing energy usage and reducing costs. This review examines the role of reinforcement learning (RL) in the advancement of HEMS, presenting it as a powerful tool for the adaptive management of complex, real-time energy demands. This review is notable for its comprehensive examination of the applications of RL-based methods and tools in HEMS, which encompasses demand response, load scheduling, and renewable energy integration. Furthermore, the integration of RL within distributed automation and Internet of Things (IoT) frameworks is emphasized in the review as a means of facilitating autonomous, data-driven control. Despite the considerable potential of this approach, the authors identify a number of challenges that require further investigation, including the need for robust data security and scalable solutions. It is recommended that future research place greater emphasis on real applications and case studies, with the objective of bridging the gap between theoretical models and practical implementations. The objective is to achieve resilient and secure energy management in residential and prosumers buildings, in particular within local microgrids.

The paper describes the concept of Human Centric Lighting (HCL, or Human Centric Lighting), and the application of modular lighting according to this concept as part of a building automation system. The process of designing and prototyping luminaires was discussed, taking into account advanced features that adapt the lighting to the ambient conditions in the room and the current needs of the users. The work demonstrated the potential of moving light sources to precisely illuminate specific areas and the importance of control algorithms to fully realize this potential. A three-layer system architecture (object, local control and supervisory) was presented, allowing parameters to be adjusted according to user preferences and optimizing operation according to criteria such as energy efficiency. The theoretical foundations based on the analysis of scientific articles and the actual technical process of building the lighting module, as well as the results of photometric tests of module prototypes are presented in the paper.

This paper presents the concept of an innovative control of a central heating system in a multifamily building based on the original thermodynamic model, the resulting architecture of the control system, and the originally designed and manufactured wireless temperature sensors for thermal zones. The novelty of this solution is the developed layers of the control system: distributed measurement and correction analysis, which is based on the existing infrastructure and the local HVAC controller. This approach allows for the effective use of the measured temperature data from thermal zones and finally sending the value of the calculated correction of settings to the controller. Moreover, in the analytical layer, a model was also implemented that calculates the necessary amount of energy based on data from the subsystem of temperature sensors located in the thermal zones of the building. The use of the algorithmic strategy presented in this paper extends the functionality and significantly improves the energy efficiency of the existing, classic, reference heating control algorithm by implementing additional control loops. Additionally, it enables integration with demand-side response systems. The presented concept was successfully tested, achieving real energy savings for heating by 12%. These results are described in a case-study format. The authors believe that this concept can be used in other buildings and thus will have a positive impact on the energy savings used to maintain thermal comfort in buildings and significantly reduce CO₂ emissions.